In previous in vitro studies, we have shown that short term in vitro incubation of murine splenocytes or cancer patient lymphocytes in lymphokines can generate cells highly lytic for fresh syngeneic or autologous tumor but not normal cells. We then used the B16 melanoma pulmonary metastasis model and virulent P815 mastocytoma model to test the in vivo antitumor efficacy of adoptive transfer of these lymphokine activated killer (LAK) cells. 100 million LAK cells, infused intravenously into mice with established tumor, were capable of curing mice in the P815 model and markedly decreasing the number of metastases and increasing survival time in the B16 model. These effects were seen whether the LAK cells were generated from splenocytes of normal mice or tumor bearing mice. The antitumor effectors were Thy 1+ and the lymphokine necessary for activation in the B16 model appeared to be interleukin-2. The investigations proposed in this paper will extend these results to tumor models of other histological types and strains of mice. In the models yielding positive effects, the dose response, injection into other anatomic compartments with appropriate homing measurements and augmentation of subsequent chemotherapy effects will be studied. The Ly phenotype of the donor effector cells and the cellular characteristics of the host response will be investigated and the lymphokines necessary for activation in vitro and augmentation of response in vivo will be studied. The use of IL-2 activated cells may provide a valuable method for the adoptive therapy of human neoplasms as well as for probing the biological role of such activated cells.